{
  "id": 15677,
  "url": "https://arxiv.org/abs/2607.29169v1",
  "title": "ActFovea: Runtime Safeguarding for VLA Policies via Spatiotemporal Visual-Action Consistency",
  "summary": "Vision-language-action (VLA) policies achieve strong performance in robotic manipulation but remain vulnerable to runtime disturbances that break the temporal alignment among visual observations, robot states, and executed actions. We introduce ActFovea, a plug-and-play safeguarding framework that detects and mitigates such failures without retraining or modifying the underlying VLA policy. ActFovea uses robot kinematics, proprioceptive states, and recent actions to construct action-conditioned",
  "authors": "Wenda Yu, Tianshi Wang, Fengling Li, Xin Li, Jingjing Li, Lei Zhu",
  "category": "research",
  "topics": "regulation,safety-alignment,agents-autonomy",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-31T08:47:57.000Z",
  "fetched_at": "2026-08-03T05:10:47.622Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/15677",
  "original_url": "https://arxiv.org/abs/2607.29169v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}